Successive Interference Cancellation

Description

Successive Interference Cancellation (SIC) is an advanced physical layer signal processing algorithm used in the receiver to separate multiple overlapping data streams transmitted simultaneously on the same time-frequency resource. It operates on the principle of leveraging power differences between user signals. In a typical SIC scenario, such as in power-domain Non-Orthogonal Multiple Access (NOMA), multiple users are multiplexed with different power levels. The receiver architecture for SIC includes multiple stages of decoding, regeneration, and subtraction.

The process works sequentially. First, the receiver decodes the signal of the strongest user (the one with the highest received power), treating all other users' signals as noise. This is possible because the high signal-to-interference-plus-noise ratio (SINR) of the strong user allows for reliable decoding even in the presence of interference. Once decoded, the receiver perfectly reconstructs (re-encodes and re-modulates) the estimated signal of this strong user, including its channel effects. This reconstructed signal is then subtracted from the original composite received signal. This cancellation step removes a significant portion of the interference for the remaining, weaker users.

After subtraction, the residual signal contains the data of the remaining users, but with the dominant interference from the strongest user largely eliminated. The receiver then moves to the next strongest user in the residual signal, repeating the process: decode, reconstruct, and subtract. This continues iteratively until all intended user signals are decoded. The key to SIC's success is accurate channel estimation and perfect signal reconstruction; any error in decoding or reconstructing an early user's signal propagates to all subsequent users, a phenomenon known as error propagation. SIC is computationally more complex than linear detection methods like MMSE but offers vastly superior performance in high-interference, capacity-constrained environments. In 3GPP, it is a cornerstone receiver technique for uplink NOMA schemes studied for massive machine-type communications (mMTC) and is also relevant for advanced MIMO detection.

Purpose & Motivation

SIC exists to overcome the fundamental capacity limitations of orthogonal multiple access schemes like OFDMA, where resources are divided among users without overlap. Orthogonality avoids interference but can be spectrally inefficient when users have diverse channel conditions or small data packets. SIC enables non-orthogonal multiplexing, allowing multiple users to share the same resource block, thereby increasing the number of simultaneous connections and the overall system throughput—a critical goal for 5G and beyond.

The historical context lies in information theory, with the concept of superposition coding and dirty paper coding establishing the theoretical benefits of non-orthogonal access. Practical implementations became feasible with increased receiver processing power. SIC addresses the problem of multi-user interference head-on. In scenarios like the uplink of a cellular network (e.g., for IoT devices), many devices transmit sporadically with small payloads. Grant-based orthogonal access would incur excessive signaling overhead. Grant-free NOMA with SIC allows devices to transmit without explicit scheduling, and the base station uses SIC to separate the collided transmissions, dramatically reducing latency and signaling.

It specifically solves the near-far problem in power-domain multiplexing. By allocating more power to users with poor channel conditions (far users) and less power to users with good conditions (near users), the base station can decode the strong far-user signal first and then the near-user signal after cancellation. This ensures fairness and coverage. Without SIC, such non-orthogonal transmissions would result in catastrophic interference and be unusable. Thus, SIC is a key enabler for massive connectivity and ultra-reliable low-latency communications where resource efficiency and low overhead are paramount.

Key Features

• Iterative decode-and-subtract process for multi-user signal separation
• Exploits power differences between users for ordered decoding
• Core enabling technology for power-domain Non-Orthogonal Multiple Access (NOMA)
• Improves spectral efficiency and system capacity in interference-limited channels
• Requires accurate channel estimation and signal reconstruction to mitigate error propagation
• Increases receiver computational complexity compared to linear detectors

Evolution Across Releases

Defining Specifications